Structural and functional changes in neural networks of attention and goal-directed cognition likely contribute to age-related memory decline and impede daily living. While considerable progress has been made in specifying how changes in the medial temporal lobe affect memory, moment-to-moment and individual differences in attention and goal-state representation are also hypothesized to impact episodic encoding and retrieval in young and older adults, and to contribute to age-related memory change. Of equal importance, in asymptomatic (`healthy') older adults, preclinical Alzheimer's disease (AD) pathology may disrupt attention and goal coding, with deleterious consequences for memory. Here, we aim to use innovative functional, molecular, and structural measures to characterize interactions between attention, goal states, and memory, and to examine (a) their contributions to trial-level, subject-level, and group-level memory differences, and (b) their relation to AD pathology. We will leverage goal-directed episodic encoding and retrieval paradigms and cutting-edge cognitive neuroscience tools, including task-based EEG-fMRI with pupillometry in asymptomatic (`healthy') older (65-79 yrs) and young adults (18-30 yrs). Trial-level attentional lapses will be assayed via fluctuations of alpha and theta oscillatory power and pupil diameter; the strength of trial-level goal states will be quantified via multivariate analyses of frontoparietal BOLD patterns. Aims 1-2 will address: How do interactions in attention and goal-state representation affect cortical and hippocampal mechanisms of episodic encoding (Aim 1) and retrieval (Aim 2), and how do age-related changes in these interactions relate to memory differences across age? Moreover, we will examine how molecular and structural biomarkers of pathological aging (AD) in `healthy' older adults relate to neural, pupillometry, and behavioral assays of attention, goal states, memory, and their interactions. Via PET-MR, we will measure (a) global ?-amyloid (A?) burden and regional A? in frontoparietal cortex, and (b) locus coeruleus integrity, a core structure for attention, arousal, and goal-directed cognition, and an early site of AD pathology. Using categorical (A?+ vs. A?-) and continuous analyses, Aim 3 will address: Do molecular and structural biomarkers of pathology in preclinical aging predict differences in attentional lapses and goal coding, accounting for significant shared or unique variance in behavioral and neural measures of memory in asymptomatic individuals? The promise, feasibility, and novelty of the proposed research are grounded in strong preliminary data and derive from the use of multi-modal measures to discover how function, structure, and early pathology interact to affect attention, goal coding, and memory in aging. The project will advance understanding of (a) how moment-to- moment and individual differences in attention and goal coding affect learning and remembering in young and older adults, and (b) how these differences relate to memory decline in aging with and without AD pathology. The latter holds promise for revealing novel neurocognitive biomarkers of AD risk.